The invention relates to a process for preparing formic acid by reaction of carbon dioxide with hydrogen in a hydrogenation reactor in the presence of a catalyst comprising an element of group 8, 9 or 10 of the Periodic Table, a tertiary amine and a polar solvent to form formic acid/amine adducts as intermediates which are subsequently thermally dissociated.
Adducts of formic acid and tertiary amines can be thermally dissociated into free formic acid and tertiary amine and therefore serve as intermediates in the preparation of formic acid.
Formic acid is an important and versatile product. It is used, for example, for acidification in the production of animal feeds, as preservative, as disinfectant, as auxiliary in the textile and leather industry, as a mixture with its salts for deicing aircraft and runways and also as synthetic building block in the chemical industry.
The abovementioned adducts of formic acid and tertiary amines can be prepared in various ways, for example (i) by direct reaction of tertiary amine with formic acid, (ii) by hydrolysis of methyl formate to form formic acid in the presence of the tertiary amine or (iii) by catalytic hydration of carbon monoxide or hydrogenation of carbon dioxide to form formic acid in the presence of the tertiary amine. The latter process of catalytic hydrogenation of carbon dioxide has the particular advantage that carbon dioxide is available in large quantities and is flexible in terms of source.
An industrially promising process appears to be, in particular, the catalytic hydrogenation of carbon dioxide in the presence of amines (W. Leitner, Angewandte Chemie 1995, 107, pages 2391 to 2405; P. G. Jessop, T. Ikariya, R. Noyori, Chemical Reviews 1995, 95, pages 259 to 272). The adducts of formic acid and amines formed here can be thermally dissociated into formic acid and the amine used, which can be recirculated to the hydrogenation.
The catalyst necessary for the reaction comprises one or more elements from group 8, 9 or 10 of the Periodic Table, i.e. Fe, Co, Ni, Ru, Rh, Pd, Os, Ir and/or Pt. The catalyst preferably comprises Ru, Rh, Pd, Os, Ir and/or Pt, particularly preferably Ru, Rh and/or Pd and very particularly preferably Ru.
To make an economical process possible, the catalyst used has to be separated off ideally completely from the product stream and recirculated to the reaction, for two reasons:    (1) large losses of the expensive catalyst would incur considerable additional costs and would be prohibitive for economical operation of the process.    (2) in the thermal dissociation of the formic acid/amine adducts, very little catalyst should be present since in the absence of a CO2 and/or H2 pressure this also catalyzes the reverse reaction and thus leads to losses of the formic acid formed.

Decomposition of formic acid/amine adducts in the presence of a catalyst (x=0.4−3)
The transition metal-catalyzed decomposition of formic acid has been described comprehensively, especially recently: C. Fellay, N. Yan, P. J. Dyson, G. Laurenczy Chem. Eur. J. 2009, 15, 3752-3760; C. Fellay, P. J. Dyson, G. Laurenczy Angew. Chem. 2008, 120, 4030-4032; B. Loges, A. Boddien, H. Junge, M. Beller Angew. Chem. 2008, 120, 4026-4029; F. Joó ChemSusChem 2008, 1, 805-808; S. Enthaler ChemSusChem 2008, 1, 801-804; S. Fukuzumi, T. Kobayashi, T. Suenobu ChemSusChem 2008, 1, 827-834; A. Boddien, B. Loges, H. Junge, M. Beller ChemSusChem 2008, 1, 751-758.
The catalysts used here are in principle also suitable in principle for the hydrogenation of CO2 to formic acid (P. G. Jessop, T. Ikariya, R. Noyori Chem. Rev. 1995, 95, 259-272; P. G. Jessop, F. Joó, C. C. Tai Coord. Chem. Rev. 2004, 248, 2425-2442; P. G. Jessop, Homogeneous Hydrogenation of Carbon Dioxide, in: The Handbook of Homogeneous Hydrogenation, Hrsg.: J. G. de Vries, C. J. Elsevier, Volume 1, 2007, Wiley-VCH, pp. 489-511). Thus, the hydrogenation catalysts have to be separated off before the thermal dissociation in order to prevent the undesirable decomposition of formic acid.
WO 2008/116,799 discloses a process for the hydrogenation of carbon dioxide in the presence of a catalyst which comprises a transition metal of transition group VIII (groups 8, 9, 10) and is suspended or homogeneously dissolved in a solution, a tertiary amine having at least one hydroxyl group and a polar solvent to form an adduct of formic acid and the tertiary amine. The hydroxyl group(s) in the tertiary amine enable an increased carbon dioxide solubility compared to the triethylamine which is usually used to be achieved. As preferred homogeneous catalysts, mention may be made of RuH2L4 having monodentate phosphorus-based ligands L and RuH2(LL)2 having bidentate phosphorus-based ligands LL and particularly preferably RuH2[P(C6H5)3]4. As polar solvents, mention may be made of alcohols, ethers, sulfolanes, dimethyl sulfoxide and amides whose boiling point at atmospheric pressure is at least 5° C. above that of formic acid. The tertiary amines which are preferably to be used also have a boiling point above that of formic acid. Since no phase separation takes place, the work-up of the entire reaction product mixture is carried out by distillation, optionally after prior removal of the catalyst, in which the adduct of formic acid and the tertiary amine which is formed is thermally dissociated and the formic acid liberated is isolated as overhead product. The bottom product comprising tertiary amine, polar solvent and optionally catalyst is recirculated to the hydrogenation stage.
A disadvantage of this process is the introduction of the entire liquid reaction product mixture into the apparatus for thermal dissociation and distillation, optionally after prior specific removal of the homogeneous catalyst by means of a separate process step, for example an extraction, adsorption or ultrafiltration step. The apparatus for the thermal dissociation and distillation consequently has to be made larger and more complex both in terms of the higher liquid throughput and the more specific separation properties, which is reflected, inter alia, in the capital costs (for example via engineering input, material, space requirement). In addition, the higher liquid throughput also results in a higher energy usage.
However, the fundamental work on the catalytic hydrogenation of carbon dioxide to form formic acid was carried out as early as the 1970s and 1980s. The processes of BP Chemicals Ltd filed as the patents EP 0 095 321 A, EP 0 151 510 A and EP 0 181 078 A may be considered to result therefrom. All three documents describe the hydrogenation of carbon dioxide in the presence of a homogeneous catalyst comprising a transition metal of transition group VIII (groups 8, 9, 10), a tertiary amine and a polar solvent to form an adduct of formic acid and the tertiary amine. As preferred homogeneous catalysts, EP 0 095 321 A and EP 0 181 078 A in each case make mention of ruthenium-based carbonyl-, halide- and/or triphenylphosphine-comprising complex catalysts and EP 0 151 510 A mentions rhodium-phosphine complexes. Preferred tertiary amines are C1-C10-trialkylamines, in particular the short-chain C1-C4-trialkylamines, and also cyclic and/or bridged amines such as 1,8-diazabicyclo[5.4.0]undec-7-ene, 1,4-diazabicyclo[2.2.2]octane, pyridine or picolines. The hydrogenation is carried out at a carbon dioxide partial pressure of up to 6 MPa (60 bar), a hydrogen partial pressure of up to 25 MPa (250 bar) and a temperature from about room temperature to 200° C.
EP 0 095 321 A and EP 0 151 510 A teach the use of an alcohol as polar solvent. However, since primary alcohols tend to form formic esters (organic formates), secondary alcohols, in particular isopropanol, are preferred. In addition, the presence of water is described as advantageous. According to the examples in EP 0 095 321 A, the reaction product mixture is worked up by directly subsequent two-stage distillation in which the low boilers alcohol, water, tertiary amine are separated off in the first stage and the adduct of formic acid and the tertiary amine is separated off at the top under vacuum conditions in the second stage. EP 0 151 510 A likewise teaches a work-up by distillation, but with reference to EP 0 126 524 A with subsequent replacement of the tertiary amine in the adduct which has been separated off by distillation by a weaker, less volatile nitrogen base before thermal cleavage of the adduct in order to aid or make possible the subsequent thermal dissociation to produce free formic acid.
EP 0 181 078 A teaches the targeted choice of the polar solvent on the basis of three essential criteria which have to be fulfilled at the same time:    (i) the homogeneous catalyst has to be soluble in the polar solvent;    (ii) the polar solvent must not have an adverse effect on the hydrogenation; and    (iii) the adduct of formic acid and the tertiary amine which is formed should be able to be readily separated off from the polar solvent.
As particularly suitable polar solvents, mention is made of various glycols and phenyipropanols.
According to the teaching of EP 0 181 078 A, the work-up of the reaction product mixture is carried out by firstly separating off the gaseous components (in particular unreacted starting materials hydrogen and carbon dioxide) at the top of an evaporator and separating off the homogeneous catalyst dissolved in the polar solvent at the bottom and recirculating them to the hydrogenation stage. The adduct of formic acid and the tertiary amine is subsequently separated off from the remaining liquid phase comprising the adduct of formic acid and the tertiary amine, free tertiary amine and possibly water and the remaining part of the liquid phase comprising the free tertiary amine and possibly water is recirculated to the hydrogenation stage. The separation can be effected by distillation or phase separation of the two-phase system (decantation).
A further significant teaching of EP 0 181 078 A is the subsequent, absolutely necessary replacement of the tertiary amine in the adduct which has been separated off by a weaker, less volatile nitrogen base before the adduct is thermally dissociated in order to aid or make possible the subsequent thermal dissociation to produce free formic acid. As particularly suitable weaker nitrogen bases, mention is made of imidazole derivatives such as 1-n-butylimidazole.
A disadvantage of the process of EP 0 181 078 A is the very complicated, four-stage work-up of the reaction product mixture by    (i) separating off the gaseous components and also the homogeneous catalyst and the polar solvent in an evaporator and recirculating them to the hydrogenation stage;    (ii) separating off the adduct of formic acid and the tertiary amine in a distillation column or a phase separator and recirculating the remaining liquid stream to the hydrogenation stage;    (iii) replacing the tertiary amine in the adduct of formic acid and the tertiary amine by a weaker, less volatile nitrogen base in a reaction vessel having a superposed distillation column and recirculating the tertiary amine liberated to the hydrogenation stage; and    (iv) thermally dissociating the adduct of formic acid and the weaker nitrogen base and recirculating the weaker nitrogen base liberated to the base replacement stage.
A further, important disadvantage of the process of EP 0 181 078 A and also of the processes of EP 0 095 321 A and EP 0 151 510 A is the fact that the adduct of formic acid and the tertiary amine partly redissociates into carbon dioxide and hydrogen in the presence of the homogeneous catalyst during the work-up in the evaporator. As a solution to this problem, EP 0 329 337 A proposes the addition of a decomposition inhibitor which reversibly inhibits the homogeneous catalyst. As preferred decomposition inhibitors, mention is made of carbon monoxide and oxidants. However, disadvantages of this are the introduction of further substances into the overall process and the necessity of reactivating the inhibited homogeneous catalyst before it is used further.
EP 0 357 243 A, too, addresses the disadvantage of the partial redissociation of the adduct of formic acid and the tertiary amine in the process of EP 0 181 078 A by joint work-up of the reaction product mixture in the evaporator. The process proposed in EP 0 357 243 A teaches the use of a homogeneous catalyst comprising a transition metal of transition group VIII (groups 8, 9, 10), a tertiary amine and two different solvents, namely a nonpolar, inert solvent and a polar, inert solvent, which form two immiscible liquid phases in the catalytic hydrogenation of carbon dioxide to form an adduct of formic acid and tertiary amine. As nonpolar solvents, mention is made of aliphatic and aromatic hydrocarbons but also of phosphines having aliphatic and/or aromatic hydrocarbon radicals. Polar solvents mentioned are water, glycerol, alcohols, polyols, sulfolanes and mixtures thereof, with water being preferred. The homogeneous catalyst dissolves in the nonpolar solvent and the adduct of formic acid and tertiary amine dissolves in the polar solvent. After the reaction is complete, the two liquid phases are separated, for example by decantation, and the nonpolar phase comprising the homogeneous catalyst and the nonpolar solvent is recirculated to the hydrogenation stage. The polar phase comprising the adduct of formic acid and tertiary amine and the polar solvent is then subjected to an absolutely necessary replacement of the tertiary amine in the adduct by a weaker, less volatile nitrogen base before thermal dissociation of the adduct in order to aid or make possible the subsequent thermal dissociation to produce free formic acid. In a manner analogous to EP 0 181 078 A, imidazole derivatives such as 1-n-butylimidazole are also mentioned here as particularly suitable weaker nitrogen bases.
A disadvantage of the process of EP 0 357 243 A is the very complicated, three-stage work-up of the reaction product mixture by    (i) separating the two liquid phases and recirculating the phase comprising the homogeneous catalyst and the nonpolar solvent to the hydrogenation stage;    (ii) replacing the tertiary amine in the adduct of formic acid and the tertiary amine of the other phase by a weaker, less volatile nitrogen base in a reaction vessel with superposed distillation column and recirculating the tertiary amine liberated to the hydrogenation stage; and    (iii) thermally dissociating the adduct of formic acid and the weaker nitrogen base and recirculating the weaker nitrogen base liberated to the base replacement stage.
A further disadvantage of the process of EP 0 357 243 A is the use of two solvents and thus introduction of a further substance into the overall process.
As an alternative, EP 0 357 243 A also discloses the possibility of using only one solvent. In this case, the addition of the polar solvent in which the adduct of formic acid and the tertiary amine would otherwise dissolve is omitted. The sole solvent used here is the nonpolar solvent which dissolves the homogeneous catalyst. However, this alternative also has the disadvantage of the very complicated, three-stage work-up as described above.
DE 44 31 233 A likewise describes the hydrogenation of carbon dioxide in the presence of a catalyst comprising a transition metal of transition group VIII (groups 8, 9, 10), a tertiary amine and a polar solvent and water to form an adduct of formic acid and the tertiary amine, in which, however, the catalyst is present in heterogeneous form and the active component is applied to a inert support. Preferred tertiary amines are C1-C8-trialkylamines, polyamines having from 2 to 5 amino groups, aromatic nitrogen heterocycles such as pyridine or N-methylimidazole and also cyclic and/or bridged amines such as N-methylpiperidine, 1,8-diazabicyclo[5.4.0]undec-7-ene or 1,4-diazabicyclo[2.2.2]octane. As suitable polar solvents, mention is made of the low-boiling C1-C4-monoalcohols, and, in a manner analogous to EP 0 095 321 A, secondary alcohols are preferred. The hydrogenation is carried out at a total pressure of from 4 to 20 MPa (from 40 to 200 bar) and a temperature of from 50 to 200° C. For the work-up of the adduct of formic acid and tertiary amine which is formed, DE 44 31 233 A teaches the use of known methods with explicit reference to the work-up with replacement of the tertiary amine in the adduct of formic acid and the tertiary amine by a weaker, less volatile nitrogen base as disclosed in EP 0 357 243 A. In a manner analogous to the process of EP 0 357 243 A, the process of DE 44 31 233 A also has the disadvantage of the very complicated, three-stage work-up of the reaction product mixture.